"K7ITM" wrote in
ps.com:

Hi Owen,

I had a quick look at your article. Though I didn't try to proof-read
it for accuracy, I was reminded that the equations I posted those long
years ago said that if you know the _instantaneous_ voltage and
current at a point on a line, and know its impedance (as a frequency-
independent quantity), the equations apply, and you can resolve that
instantaneous pair of values into forward and reverse. That's
something that's not immediately obvious when people think only about
sine waves.

Tom,

Something else that follows from the derivation is that whilst the
indicated Pf and Pr do not have stand along meaning, Pf-Pr does have
meaning irrespective of the nominal R for which the instrument is
calculated.

For example, if we cascade a 100W source, 50 ohm directional wattmeter, a
75 ohm directional wattmeter and a 100+j0 load, the instrument readings
should be:
- 75 ohm: Pf=112.4, Pr=12.4, P=100
- 50 ohm: Pf=104.1, Pr=4.1, P=100

This of course assumes that the instruments do not significantly disturb
the thing they are measuring, in this case the V/I conditions at the 100
ohm load.

So, while you can nominate any reference Zo for a Pf or Pr value (and so
vary those values), the power passing the instrument (Pf-Pr) is indicated
correctly irrespective of the calibration R.

(The article explains that the result of Pf-Pr is only meaningful if the
calibration impedance is purely real.)

Owen